William Osler.

Modern medicine : its theory and practice, in original contributions by American and foreign authors (Volume v. 5) online

. (page 71 of 126)
Online LibraryWilliam OslerModern medicine : its theory and practice, in original contributions by American and foreign authors (Volume v. 5) → online text (page 71 of 126)
Font size
QR-code for this ebook

after the rupture was in communication with a bile duct. The abdomen
usually enlarges slowly, but it may do so rapidly. There is no fever, and it
is remarkable that, in spite of the marked absorptive powers of the peritoneum,
jaundice does not occur. The effusion has a great tendency to recur, and
several tappings may be required.


Although acute peritonitis means acute inflammation of the peritoneum,
the term is used to describe cases showing the clinical effects of acute in-
fection of the peritoneum, namely, septicaemia and toxsemia, in which there
may be very little or no manifest inflammatory reaction on the part of the
peritoneum. These cases are more suitably described as "acute peritoneal
infection." Thus, although it may sound paradoxical, the most severe
cases of peritoneal infection often show the least evidence of inflammatory
reaction. Peritonitis is indeed a salutary process in so far that it prevents
excessive absorption from the peritoneum, leads to destruction of micro-
organisms that have gained entrance into the cavity of the abdomen, and
by the formation of fibrin and adhesions prevents the spread of infection
throughout the peritoneal cavity.

Acute diffuse peritonitis is not necessarily universal and co-extensive with
the whole area of the peritoneum. Strictly speaking, cases of acute peri-
tonitis might be divided into (1) circumscribed, (2) diffuse or spreading,
and (3) universal or total. But in practice acute peritonitis is usually
considered as either (1) circumscribed or localized, or (2) general or

There is some unavoidable vagueness about the terms general and diffuse
as descriptive of the extent of the peritonitis seen at an operation; thus,
the word employed may be meant to imply universal infection and inflam-
mation of the peritoneum throughout the whole of the abdomen, or only
that at the time of the operation the peritonitis was not strictly localized.
It is therefore somewhat difficult to judge of the value of the statistical
results in cases of peritonitis described as diffuse or general. It is, however,
usually impossible, and certainly inadvisable, to determine during an opera-
tion whether the peritonitis is absolutely universal or only diffuse, and in
any case universal peritonitis is a later stage of the diffuse form. It has,
therefore, been suggested that the term universal should be employed instead
of general, and progressive instead of diffuse. The term diffuse will be
employed here.

1 Rev. de chir., Paris, 1902, xxvi, 67. ^ Qaz. des hop., Paris, 1906, 1065.

VOL. V. — 33


Pathogeny. — Since chemifal poisons, such as turpentine, croton oil,
antl nitrate of silver, when introchiced into the peritoneal cavity of animals,
can undoubtedly set up acute inflaniination, the question arises whether
general peritonitis in man is ever due to toxins in the absence of bacteria.
In other words, whether chemical peritonitis in contradistinction to bacterial
peritonitis occurs in man. In cases of bacterial infection of the vermiform
appendix or gall-bladiler local peritonitis in the immediate neio;hborhood
may conceivably be due to diffusion of toxins through the walls of the in-
flamed viscera. Cultures and coverslip preparations from cases of this
kind in an early stage may, indeed, show an absence of microorganisms.
This inflammatory reaction would, however, be strictly localized, and
in any case after a short time microorganisms would pass tlirough the
damaged walls of the affected viscera.

Sterile bile and urine do not set up peritonitis, and although trypsin may
produce a hemorrhagic inflammation, the secretions of the pancreas and
stomach are, when they gain entrance into the peritoneal cavity, practically
always mixed with the contents of the alimentary canal, and therefore with
bacteria. In cases in which corrosive poisons are taken by the mouth into
the alimentary canal, the damage to the walls of the stomach and intestines
would enable bacteria to pass into the peritoneum. Moreover, in inflam-
mation of the peritoneum set up by chemical means, bacteria would almost
certainly gain an entrance sooner or later from the alimentary canal or by
the blood stream. It is true that cases of peritonitis occur in which cultures
from the peritoneum prove sterile, but Durham^ has shown how rapidly
microorganisms are removed from the peritoneum, and that in cases of
peritonitis in which the exudation is sterile the omentum contains bacteria.
Whilst, therefore, the possibility of chemical (aseptic) peritonitis may be
admitted, it is safe to assume that in practice all cases of acute peritonitis
are due to bacterial infection.

Although acute diffuse peritonitis is from a practical point of view always
due to bacterial infection of the peritoneum, the mere entry of microorganisms
into the peritoneal cavity is not alone enough to produce peritonitis. The
defensive powers of the peritoneum, viz., the phagocytic action of the endo-
thelial and other cells, the bactericidal power of the peritoneal fluid depend-
ing on the presence of antibodies, and absorption^which destroy and
remove the invading bacteria — are able to prevent the production of inflam-
mation. With regard to absorption from the peritoneal cavity, it should be
pointed out that, while the removal of bacteria in comparatively small
cjuantities and their destruction in the lymphatic glands and other organs
prevent the onset of peritonitis, and are therefore beneficial, the passage
of large numbers of virulent bacteria into the general circulation, which
occurs when the endothelial cells of the peritoneum are damaged, gives rise
to a grave and often fatal bacterisemia. Death from "shock" after perfora-
tion may in all probability be explained in this way. The absorptive powers
of the peritoneum also militate against peritonitis by removing fluid in which
microorganisms might multiply. The defensive powers of the peritoneum
have been divided by Andrewes^ into those w^hich are physiological, or
the "first line of defence," and those which come into play when inflam-

1 Journal of Pathology and Bacteriology, London and Edinburgh, 1897, iv, 357.
^ General Pathology and Bacteriology of Acute Peritonitis. System of Medicine
(Allbutt and RoUeston), 1907, vol. iii, 897.


mation is set up, or the "second line of defence." These two lines of defence
merge into each other, for example, in phagocytosis, but the formation of
fibrinous adhesions which localize the infective focus is a frankly inflam-
matory process.

When the defensive powers of the peritoneum are inhibited fjr prevented
from exerting their full effects, bacterial invasion is enabled to set up peri-
tonitis. Factors which reduce the resistance of the peritoneum, such as the
sudden entrance of poisonous fluids, drying or traumatism during the course
of an operation, and possibly cold, may interfere with the a})sorptive and
bactericidal powers of the peritoneum. Again, the presence of sf>lid bodies,
such as particles of food, fseceS, blood clot, concretions, or foreign bodies,
or of ascitic fluid, may protect the bacteria, provide a nidus for their multi-
plication, inhibit absorption, and so enable them to produce their toxins
and thus set up inflammation of the peritoneum.

Etiology. — Acute diffuse peritonitis may be (1) primary or (2) secondary.

1. The term primary idiopathic peritonitis has been employed in two senses.
Formerly it was used to imply that inflammation attacked the peritoneum
alone, the rest of the body being healthy, and that its cause and origin were
unknown. Cases of peritonitis formerly thought to be due to rheumatism
and to cold came under this heading. More recently, since it has been recog-
nized that acute peritonitis is practically always infective, the term primary
peritonitis has been applied by Flexner^ to cases in which, as there is no local
focus in the abdomen to account for the infection of the peritoneum, it is
assumed that microorganisms have reached the peritoneum by the blood
or lymph stream. Hsematogenous peritonitis occurs as a terminal infection
in chronic nephritis, arteriosclerosis, cancer, and other conditions which
reduce the bactericidal power of the blood. Some cases of peritonitis com-
plicating erysipelas or other infections, and about half of the instances
of pneumococcal peritonitis, would also appear to be primary and so of
hsematogenous origin. Since a minute local focus, such as a microscopic
abscess in the appendix, may be easily overlooked, a most exhaustive post-
mortem examination is necessary before a case of primary idiopathic peri-
tonitis can be accepted, and many writers are somewhat unwilling to admit
its existence. Of this, however, there can be no doubt, and probably about
10 per cent, of the fatal cases of general peritonitis are of this nature. It
occurred in 12 out of 106 cases examined by Flexner and in 9 out of 105 cases
at the Massachusetts General Hospital (Manahan^). The infection is
usually single; thus out of Flexner's 12 cases no bacteria were found in 2,
in 9 there was a single infection, and in 1 a mixed infection.

2. Secondary 'peritonitis, in which infection is due to some local lesion
either in or in the immediate neighborhood of the abdomen, accounts for
the great majority of cases of acute diffuse peritonitis.

(a) Wound or -operation introducing infection into the peritoneal cavity,
e. g., after herniotomy. To this group Flexner applies the term exogenous.
But it must, of course, be remembered that postoperative peritonitis may
be due to bacterial infection derived from the viscera.

(b) Due to infection derived from the abdominal viscera. This may
depend on gross perforation of their walls or be due to inflammation allowing

* Philadelphia Medical Journal, 1902, ii, 1019.

^ Boston Medical and Surgical Journal, 1905, clii, 345.


micToor<2;ani.sm.s to pass tliroujjli the intestinal walls. This is Flexncr's
endogenous groiij). It will be most convenient to give a list of ali'eetions
of the abdominal viscera which may give rise to acute peritonitis.

Stomach. — Perforation of a gastric ulcer; perforation due to rapidly
growing, soft, and necrosing carcinoma; acute suppurative (or phlegmonous)
gastritis; acute gastritis due to corrosive poisons.

Small Intcsfincs. — Traumatic ru])ture, perforation by sharp foreign bodies;
perforating ulcers, duodenal; jejunal ])cptic ulcer, occurring in cases of
gastrojejunostomy, esjiecially, it is said, after the anterior o])eration; typhoid
ulcer (peritonitis in rare instances occurs in the al)sence of any gross per-
foration); unemic ulcer; acute enteritis; strangulation, kinking, volvulus,
intussusception; internal hernia; infarction.

Appendix. — Perforation as a result of acute infective inflammation;
from rupture of an abscess in the walls of the appendix or in its neighborhood,
or from acute appendicitis as already mentioned. An abscess in the
appendix so minute as to be easily overlooked may rupture into the peri-
toneum and set up peritonitis, which may thus appear to be primary or

Colon. — Perforation, traumatic or by a foreign body from within. Per-
forating ulcer, stercoral dysenteric; in connection with carcinoma, either
from rupture of a distention ulcer, or leakage through necrosing growth.
Extremely acute inflammation, in strangulation, volvulus, intussusception.
Wounds, perforation of the rectum by a bougie, nozzle of an enema syringe.

Gall-bladder. — Acute, phlegmonous, or gangrenous cholecystitis. Rup-
ture of an inflamed gall-bladder.

Liver. — Rupture or leakage of an abscess or of a suppurating hydatid
cyst. Abscesses may be multiple and due to arterial pyemia, infection of
the portal vein (pylephlebitis), or of the bile ducts (suppurative cholangitis).
In newborn infants phlebitis of the umbilical vein gives rise to perito-

Spleen. — Rupture of an abscess, or of a suppurating infarct.

Mesenteric Glands. — Rupture of a suppurating lymphatic gland into the
general cavity of the peritoneum.

Ulceration and perforation of the urinary bladder or of sacculi in connec-
tion with it.

Suppurating and gangrenous ovarian cysts.

Pyosalpinx and infection spreading from the Fallopian tubes. The
importance of this factor has probably been overestimated in the past.
RiedeP has shown reasons for believing that salpingitis is a more important
cause of general peritonitis in girls under ten years of age than in adult

Perforation of the uterus by a sound.

Rupture of an abscess into the peritoneal cavity, especially of a peri-
appendicular abscess; this may follow rough manipulation. It is remarkable
how rarely erysipelas or suppuration in the abdominal parietes gives rise
to peritonitis; this probably depends on the lymphatics being independent.
Although peritonitis may lead to infection of the pleurse, an empyema
hardly ever ruptures into the abdominal cavity. Peritonitis readily spreads
from the pancreas to the peritoneum of the lesser sac. Retroperitoneal

' Arch.f. klin. Chir., Berlin, 1906, Ixxxi, 186.


suppuration, such as that (hic to tuberculous ostitis of the spine or to renal
calcuU seldom infects the peritoneum. In infants acute ])eritonitis is usually
due to infection from the umbilicus.

As giving the relative importance of the various starting points of a(,-ute
peritonitis, Benda's^ 446 cases of acute diffuse peritonitis examined after
death may be referred to. The vermiform ap})endix was tlie starting point in
115 cases, the stomach and duodenum in GS, the rest of the intestine in
118, the female genitals in 81, the gall-bladder in 10, the kidney and urinary
bladder in 10, the pancreas in 2, the spleen in 1; 35 were unknown, 4 were
postoperative, and 2 were described as hsematogenous (nephritis and rheu-
matic fever).

The sex and age incidence depend on those of the commonest causes of
peritonitis, perforating gastric ulcer and pelvic infections being the most
frequent etiological factors in women, and appendicitis in children and
young males. In 200 fatal cases at St. George's Hospital there were 107
males and 93 females; the average age of the 200 cases was 32 years, being
34.26 in the males and 29.3 years in the females. The largest number of
cases occur between twenty and forty years of age.

Bacteriology. — Although a great deal of work has been done on this
subject our knowledge is far from complete. In the following account
the results obtained by Dudgeon and Sargent will be freely utilized. A
large number of microorganisms have been found in human peritonitis,
the most important of which will now be considered. In many instances
there is a mixed infection, and difficulty arises in apportioning the share that
the organisms present have taken in producing peritonitis. The organisms
found in peritonitis are in nearly all instances morphologically the same as
those commonly present in the healthy intestine.

The Bacillus coli communis is very commonly found, and considerable
discussion has taken place whether its importance is commensurate with
its frequency. It has been regarded as the all-important factor in perito-
nitis, and, on the other hand, it has been recently thought that the Bacillus
coli, being more hardy than other bacteria, may overgrow them and so be
found postmortem, to the exclusion of other microorganisms which are the
real factors in producing peritonitis. That this disappearance of the
original organism can occur has been shown experimentally by Klein, and
in man an organism, the pneumococcus, found in pure culture at operation,
has not been obtained after the death of the patient, while Bacillus coli
was (Charrin and Veillon^). But these observations do not in any way
prove that the colon bacillus is not an important cause of peritonitis. It is
naturally found most often in cases of peritonitis due to infection from the
alimentary canal, and is very frequently present in association with some
other organism; in 56 cases of peritonitis due to bowel infection, Flexner
found the Bacillus coli in 43, in pure culture in 8, and in association with
other organisms in 35. Dudgeon and Sargent,^ as the result of bacterio-
logical examination of 270 cases, came to the conclusion that the colon bacillus
is the most important factor, and that it is found in the largest number of
fatal cases. Its virulence varies greatly, and while it is less virulent than

^ Quoted in A System of Surgery, von Bergmann, von Bruns, and von Mikulicz,
translation, vol. iv, 165.

2 Compt. rend.. Soc. hiol., Paris, 1893, xlv, 1057.

^ The Bacteriology of Peritonitis, Constable & Co., London, 1905.


the streptococci and Bacillus pyorijaneus, it is more frequently present.
With mixed cuhures a very virulent ])eritonitis is set up.

A white staph iilococcus, which is not the same as tlie Staphylococcus pijogeries
alhtis descnl)ed by Roseiihach, and has often been reijarded as due to con-
tamination from the skin, and is ]M"()l)ahly Welch's StapJti/locorcus rpider-
midis alhus, has been found by Dudg-eon and Sargent to be the hrst organism
to appear in the peritoneal cavity in cases of early peritonitis, and to pass
through the walls of the inflamed intestine before more virulent organisms.
They found this microorganism in 108 out of 258 cases of ditt'erent peri-
toneal lesions.^ It is an organism of low virulence, and, as mentioned above,
gives rise to an exudation rich in phagocytes and thus exerts a protective
action against more virulent microorganisms.

Strrptococci arc found in a considerable percentage of the cases. Flexner
found them second in order of frequency both in his primary and secondary
groups of peritonitis. As streptococci are among the normal saprophytes
of the intestinal canal, it is natural that they should be commonly present
in cases of jicritonitis due to intestinal lesions, but it must be remembered
that the Streptococcus pyogenes has not been foimd normally in the alimen-
tary canal, and that streptococci found in the peritoneum are not neces-
sarily Streptococcus pyogenes, and may be the saprophytes of the intestine
(Andrewes and Horder'). Peritonitis due to Streptococcus pyogenes is the
most severe and fatal form; there is little or no local reaction, the phagocytes
being powerless against the virulent microorganisms which, as shown by
Buxton's^ experiments on animals, are rapidly absorbed into the circulation
and give rise to bacteriiemia, proving fatal in the course of 24 to 48 hours
from liberation of their endotoxins.

Staphylococcus pyogenes aureus was found by Flexner in 15 out of 34
cases of exogenous peritonitis, being in pure culture in 12 and mixed in 3.
It was present in 3 cases only in 39 cases of endogenous peritonitis. It
gives rise to a virulent form of peritonitis.

Bacillus pyocyaneus is responsible for a certain number of cases of acute
peritonitis, and ranks next to the Streptococcus pyogenes in the virulence
of the resulting peritonitis. It does not produce green pus in the peritoneal
cavity, as it often does in a local abscess. Dudgeon and Sargent have
shown that Bacillus coli will not grow in the presence of Bacillus pyocyaneus.

Pneumococci have been isolated in a small percentage of cases, but much
attention has been paid to pneumococcal peritonitis, as it is comparatively
often primary. It has been found in cases of peritonitis due to appendi-
citis, or to perforation of a gastric ulcer, peptic or malignant.

Gonococcal peritonitis is infrequent, and when it does occur is chiefly
found in women and children. No doubt some cases of gonococcal peri-
tonitis are mixed infections, and the severity of such cases is due to the other

Bacillus typhosus has in a few instances been found in cases of peritonitis
due to perforation of typhoid ulcers, and even after rupture of the spleen.
But it is combined with saprophytic organisms, and it is doubtful what
importance should be attached to its presence.

Other organisms, such as Bacillus proteus, Bacillus anthracis, are occa-

1 Lancet, 190G, ii, 1337. ^ /^^-^^ iggg^ ij^ 708.

^ Journal of Medical Research, Boston, 1907, xvi, 39.


slonally found. With regard to anaerobic organisms, there has been a good
deal of discussion; Veillon and Zuber and Tavel and Lanz conclude
that anaerobic organisms are of importance in the production of appendicitis
and peritonitis, whilst Dudgeon and Sargent employed strictly anaerobic
precautions in a number of cases of appendicitis and in other varieties of
peritonitis, but failed to find anaerobic organisms ext-ept in one case of gan-
grenous appendicitis with abscess in which Bacillus aerogenes capsulatus
was found in addition to Bacillus c.oli and a streptococcus. As the result
of the presence of anaerobic organisms the peritonitic exudate may become

Hill and Fisch^ report a case of peritonitis due to the influenza bacillus,
and refer to another case examined by von Recklinghausen. One case
occurred in the Johns Hopkins Hospital.

It may be well to summarize the data at our disposal as to the bacteriology
of peritonitis due to various anatomical lesions. In perforative peritonitis
due to gastric ulcer, the pneumococcus and a streptodiplococcus, an organ-
ism of low virulence (Dudgeon and Sargent), have been thought to be the
responsible organisms. In perforation of the small intestine the Bacillus
coli and streptococci, and in appendicitis various organisms, especially
streptococci (Low, Lartigau), Bacillus coli, and diplococci have been re-
garded by different observers to be the causal organisms. Krogius, A. O. J.
Kelly, and Dudgeon and Sargent are opposed to the view that streptococci
commonly give rise to appendicitis. In puerperal peritonitis streptococci
are most often the responsible organism.

Morbid Anatomy. — The morbid appearances seen in the abdomen
necessarily vary with the causes and stages of acute peritonitis. In the
most rapidly fatal cases, in which death is due to septicaemia from free absorp-
tion from the peritoneal cavity, there may be little or no evidence of morbid
change in the peritoneum. In other cases the earliest visible change is
congestion of the small bloodvessels of the peritoneum; this is not uniformly
distributed, but is naturally best marked and at first confined to the starting
point of the inflammation and to the omentum which is often adherent to
the focus of infection. Further it is more marked on the visceral than on
the parietal layer of the peritoneum, and is in most instances not diffuse,
but is confined to longitudinal lines which correspond with the spaces left
between contiguous and distended coils of intestine. The next change
is the exudation; at first there is slight dulling of the smooth peritoneal
surface, which appears finely granular, and on scraping yields a little fibrin.
This fibrin is sticky and as it increases in amount glues the intestines
together and collects in the triangular chinks left between adjacent coils.
At first the fibrin is thin and clear, but it becomes opaque, and may be yellow
or even of a green tint. Large masses of fibrin may eventually form and
show a somewhat spongy or reticulated structure. This fibrin serves a useful
purpose in limiting the field of infection and tending to localize the periton-
itis, and also in protecting the peritoneal endothelium and so preventing
absorption of bacteria and their toxins into the blood stream in overwhelm-
ing quantities.

The fluid portion of the exudation, usually comparatively small, may in
some instances be considerable (several quarts). It collects between the

' St. Louis Medical Review, 1903, xlviii, 55.


intestinal coils; during life it is often maiiily in the loins, but after death
it is more commonly seen in the pelvis. Its characters vary much in differ-
ent cases: it may be (1) serous; (2) serofibrinous, turbid, with pieces of
fibrin floatino; in it, and of a specific fijravity of 1018 or upward; (3) fibrinous;
(4) fibrinopurulent; (5) purulent; ((>) sanious; (7) putrid, due to the presence
of putrefactive anaerobic org-anisms, such as the Bacillus capsulatus aero-
genes. In such cases there may be free gas in the j)eritoneal cavity without
any perforation of the hollow viscera.

The character of the exudate differs in difterent forms of infection. In
the virulent streptococcal infection there is a little thin, odorless fluid; in

Online LibraryWilliam OslerModern medicine : its theory and practice, in original contributions by American and foreign authors (Volume v. 5) → online text (page 71 of 126)